test / README.md
nicofirst1's picture
Create README.md
d279555 verified
|
raw
history blame
3.56 kB
metadata
Dataset_Metadata:
  General_Information:
    Title: Concise and descriptive name of the dataset.
    Description: >-
      A brief summary explaining the dataset’s purpose, contents, and potential
      applications. Max 250 chars.
    Creator: >-
      Name of the individual or organization primarily responsible for creating
      the dataset.
    DOI_or_Other_Identifiers: >-
      A digital object identifier (DOI) or other unique identifiers if
      applicable.
    Intended_Use:
      Description: >-
        For detailed descriptions of what the dataset is intended for, including
        specific use cases envisioned by the creators (training, finetuning,
        evaluating, etc.) and instruction on how to load/store and use the data,
        follow this link:
      Link: URL to documentation or webpage.
  Access_and_Usage:
    Access_Rights: Descriptions of who can access the data and under what conditions.
    License:
      Description: >-
        Specific licensing information, select preferably from the list provided
        or use N/A if not found.
      Choices:
        - GPL
        - MIT
        - Apache 2.0
        - Creative Commons
        - N/A
    Date_of_Publication: When the dataset was made available.
    Regulatory_Compliance: >-
      Relevant compliance information if the data is subject to specific
      regulations (e.g., GDPR, HIPAA).
    Presence_of_PII:
      Description: Indicate the presence of Personally Identifiable Information (PII).
      Choices:
        - 'Yes'
        - 'No'
        - Cannot exclude
    Presence_of_copyrighted_material:
      Description: Indicate if there is any copyrighted material.
      Choices:
        - 'Yes'
        - 'No'
        - Cannot exclude
  Technical_Specifications:
    Format: The file format(s) of the dataset (e.g., CSV, JSON).
    Language: Primary language(s) of the dataset content, using ISO language codes.
    Version: Version of the dataset to track updates and changes.
    Size: The size of the dataset.
    API_Access: Information on how to access the dataset via APIs, if available.
  Content_Specifics:
    Subjects_Keywords: >-
      A set of keywords or subjects describing the dataset’s focus, which aids
      in searchability and categorization.
    Modality_ies:
      Description: Data modality included in the dataset.
      Choices:
        - Text
        - Video
        - Audio
        - Tabular
        - Multimodal
    Temporal_Coverage: Time period during which the data was collected.
    Geographical_Coverage:
      Description: >-
        Specifies the geographic area the data pertains to. Providing more
        specific details is preferable if known.
      Examples: Global, America, Asia, Europe, Italy, China, Morocco.
  Data_Quality_and_Methodology:
    Data_Collection_Method: >-
      Brief description of how the data was collected (e.g., survey, automated
      sensor).
    Data_Processing_Details: Information on how the data was processed or transformed post-collection.
    Quality_Control_Measures: Description of steps taken to ensure data quality.
    Evaluation_Protocols: >-
      Standard procedures recommended for evaluating models trained on this
      dataset, which might include specific metrics or test environments.
    Fairness_and_Bias: >-
      Information on any efforts made to ensure the dataset is balanced or to
      identify and mitigate biases present in the data.
  Provenance_and_Lineage:
    Source: Origin of the data, if derived from another dataset or source.
    Related_Datasets: >-
      Links or references to datasets that are related or used in conjunction
      with this dataset.